[Selection of statistical methods for estimating the association between exposure factors and rare outcomes based on cohort studies].

Publisher: Zhonghua yi xue hui; Zhongguo yi xue ke xue yuan Liu xing bing xue wei sheng wu xue yan jiu suo Country of Publication: China NLM ID: 8208604 Publication Model: Print Cited Medium: Print ISSN: 0254-6450 (Print) Linking ISSN: 02546450 NLM ISO Abbreviation: Zhonghua Liu Xing Bing Xue Za Zhi Subsets: MEDLINE

Publication: Beijing : Zhonghua yi xue hui; Zhongguo yi xue ke xue yuan Liu xing bing xue wei sheng wu xue yan jiu suo
Original Publication: [Beijing] : Zhonghua yi xue hui; Zhongguo yi xue ke xue yuan Liu xing bing xue wei sheng wu xue yan jiu suo, [1981-

Cohort Studies*; Humans ; Logistic Models ; Multivariate Analysis ; Regression Analysis

Odds ratio ( OR ) and relative risk ( RR ) are the most commonly used statistical indicators for the estimation of the association between exposure and outcome. In the cohort study with rare outcomes, the estimated OR approximately equals RR , but RR seems more interpretable. The study aims to explore the difference between OR and RR estimated by different multivariate analyses to provide reference for the selection of more appropriate multivariate regression methods and reporting indicators for estimating the association between exposure and rare outcome in cohort studies. This case study used the data from China birth cohort study. Modes of conception and congenital disabilities were regarded as exposure and outcome, respectively. Maternal age, family history of congenital disabilities with clear evidence were included as covariates. Logistic regression, log-binomial regression, and Poisson regression were used to estimate the OR and RR , respectively. Then, OR , RR , and their 95% CI estimated by three regression models were compared. The OR estimated by logistic regression was approximately equal to the RR estimated by log-binomial regression or Poisson regression. However, the RR estimated by log-binomial regression or Poisson regression was closer to 1.00, with a narrower 95% CI . Log-binomial regression or Poisson regression might have non convergence or over dispersion problems. It is recommended to report the RR obtained by log-binomial regression or Poisson regression in the cohort study with rare outcomes if applicable.

2016YFC1000105 National Key Research and Development Program of China; 72174128 National Natural Science Foundation of China; L222124 Beijing Natural Science Foundation; QML20231208 Beijing Hospitals Authority Youth Programme

Local Abstract: [Publisher, Chinese] 比值比（ OR） 和相对危险度（ RR ）均是评估暴露因素与研究结局间关联的常用指标，在罕见结局的队列中， OR 值常被用作 RR 值的近似估计，但 RR 值的意义更加清晰易解释。本研究旨在基于罕见结局队列研究，比较不同多因素回归模型获得 RR 与 OR 估计值的差别，为基于队列研究估计暴露因素与罕见结局间关联关系时选择多因素回归方法，以及优先报告关联大小估计指标提供参考。本研究基于中国出生队列数据开展实例研究，以全部病种的出生缺陷为研究结局，以受孕方式为暴露因素，纳入孕妇年龄、是否有出生缺陷家族史等有明确证据支持的变量作为协变量，分别拟合logistic回归、log-binomial回归以及Poisson回归，并比较 OR 和 RR 的点估计值及其95% CI 。结果表明，在罕见结局队列研究中logistic回归估计的 OR 值与log-binomial回归及Poisson回归估计的 RR 值近似，但log-binomial回归及Poisson回归估计的效应值更接近1.00，且效应值的95% CI 分布更窄，但可能存在不收敛或过离散问题。针对罕见结局的队列研究，在适用前提下，推荐优先报告基于log-binomial回归或Poisson回归获得的 RR 值。.

Date Created: 20230724 Date Completed: 20231023 Latest Revision: 20240304

20240305

10.3760/cma.j.cn112338-20230106-00011

37482717